The relationship between brokers’ influence, strength of ties and NPD project outcomes in innovation-driven horizontal networks

This paper examines the impact of a broker's perceived use of power – position (i.e., coercive, reward and legitimate) and personal (i.e., expert, information and referent) – on strength of ties between network members and new product development (NPD) project outcomes. Our sample consists of 100 individuals drawn from 42 organizations that were involved in different innovation-driven horizontal networks. The results of structural equation modeling suggest that the perceived use of both personal power bases and position power bases by the broker are positively related to the strength of ties between members. Strength of ties, on the other hand, is positively related to NPD project outcomes of design performance and development time. Finally, results show that the relationships between a broker's use of different power bases and NPD project outcomes are fully mediated by the strength of ties between networks members. Implications for research, theory, and practice are discussed.     

In search of efficient network structures: the needle in the haystack

Searching for efficient networks can prove a very difficult analytical and even computational task. In this paper, we explore the possibility of using the genetic algorithms (GA) technique to identify efficient network structures in the case of non-trivial payoff functions. The robustness of this method in predicting optimal networks is tested on the two simple stylized models introduced by Jackson and Wolinsky (1996), for which the efficient networks are known over the whole state space of the parameters’ values. This approach allows us to obtain new exploratory results in the case of the linear-spatialized connections model proposed by Johnson and Gilles (Rev Econ Des 5:273–299, 2000), for which the efficient allocation of bilateral connections is driven by contradictory forces that push either for a centralized structure around a coordinating agent, or for only locally and evenly distributed connections. Murat Yıldızoğlu gratefully acknowledges the support of the CCRDT program of Aquitaine Region.     

基于粗神经网络的港口集装箱吞吐量预测

运用粗糙集神经网络建立了广州港集装箱吞吐量的预测模型,预测了2007～2010年的集装箱吞吐量。该预测方法融合了粗糙集理论与神经网络方法具有的优点,具有很强的学习与泛化能力,非常适合处理多因素、非线性的复杂系统。预测结果对广州港的发展有较强的借鉴作用,可以为广州港未来发展提供参考。     

一种全新的两阶段个人信用评分方法探究

目前我国还没有一套规范的个人信用评分指标体系和方法。本文利用真实的个人消费信贷数据,首先建立了个人信用评分的多元线性判别分析模型和BP神经网络模型,然后将线性判别分析模型的结果与其它变量一起作为输入变量建立了混合两阶段个人信用评分模型。实证研究表明,混合两阶段个人信用评分模型相对于前两种单一模型能同时满足预测精度和稳健性的双重要求,从而,突破了通常应用单一模型于个人信用评分领域的局限。     

应用数据挖掘技术识别财务报表舞弊的方法研究

如何识别财务报表舞弊成为国内外学者普遍关注的问题。基于传统识别技术的诸多固有缺陷,近年来数据挖掘技术应用到识别财务报表舞弊领域成为发展趋势。数据挖掘技术包括统计方法、人工神经网络和其他数据挖掘技术。文章不仅对各种识别模型和技术进行了归纳,而且比较分析了各种模型和方法的优缺点、判别正确率以及适用条件,为下一步实施实证研究提供支持。     

区域耕地数量变化预测方法的对比研究

研究目的:对区域耕地数量变化预测方法进行对比分析,科学合理确定区域耕地数量。研究方法:通过Pearson相关分析法筛选影响耕地变化的关键因子,利用改进的BP神经网络算法、灰色模型和多元线性回归模型法对费县耕地数量变化进行预测。研究结果:改进的BP神经网络算法相对传统的灰色模型和多元线性回归模型等预测方法具备较高的预测精度。研究结论:改进的BP神经网络算法是进行耕地数量变化预测的较好方法,具有实际指导意义,其预测结果可以为当地相关部门合理地确定耕地保有量、推进耕地保护提供科学依据。     

The impact of external forces on cartel network dynamics: Direct research in the diamond industry

This paper describes the impact of external environmental forces on cartel networks. Using a case research approach, this report examines two leading business networks within one industry, over time. The results suggest that (a) bargaining power of intermediaries increases with the advent of new and powerful actors, (b) process activities that cartels previously controlled are being outsourced to new actors sometimes based in developing countries, (c) other actors are acquiring resources once dominated by a cartel, (d) external forces triggered by the illegal diamond trade, such as international regulatory constraints, no longer favour cartels like De Beers, and (e) over time, these and additional environment factors are forcing actors like De Beers who perform rigid process activities to become more flexible. For example, forces are moving cartels which relied previously on hand-picked intermediaries in highly controlled networks to market their products to adopt a flexible market-focused expansion of operations in retail contexts.     

基于卷积神经网络的列车检测算法研究

列车检测作为列车自动驾驶的核心技术,可以有效地降低列车追尾等事故造成的人身危险和财产损失。为实现精准的列车检测,选用改进的卷积神经网络(PVANET)对输入图像进行特征提取,在此基础上,采用候选区域网络,从生成的特征图里滑动搜索,判断出图像中可能为列车的区域位置,并进一步采用快速区域卷积神经网络对每个候选区域进行分类,计算出其所属类别的置信度,同时精确定位列车。经验证,该方法适应范围广、鲁棒性高,可以有效地检测不同环境光强及不同朝向的列车,保障列车安全,为列车自动驾驶及辅助驾驶提供安全保障。     

The Governance of Networks and Economic Power: The Nature and Impact of Subcontracting Relationships

Abstract.  Current debate on networking focuses on network structures and firm strategies. In this perspective, theoretical analysis has been concerned with allocative issues. This essay proposes a different interpretation. Starting from the existing theoretical framework, we emphasise the nature and the implications of different types of networks with respect to socio‐economic development from a distributional point of view. Within this context, we develop the analysis of subcontracting starting from the concept of economic power. We then provide an analysis of governance in production by considering the attitudes and the nature of the actors involved. The externalisation of activities by large transnationals, which characterises current corporate restructuring, is often related to the search for greater flexibility, but also for greater power over governments, labour, and subcontractors. Differently, networks based on the mutual dependence of actors, which are not necessarily built around a large firm, could – under particular conditions – reach large production scales or more complex scopes without breaking the links with territorial systems, thus including local objectives in the strategic decision‐making process. Our conclusion is that the impact of subcontracting networks varies enormously. This is crucial to an understanding of future trends and possibilities. Not least, firms and public policy agencies need to understand the implications of different forms of subcontracting network and how those forms actually differ in practice.     

生态视角下协同创新网络构建及演替研究

创新驱动已成为社会经济发展的主要驱动力,而以多元主体为基础的协同创新网络正成为创新模式的发展趋势。首先,通过对生态群落和协同创新网络的构成要素及要素间关系进行比较发现,二者具有较强相似性,因而成熟的生态群落理论可以为协同创新网络的构建和发展提供理论借鉴。然后,在生态群落多样性、层次性、协同竞争性、开放性和动态性等特征的指导下构建了协同创新网络。在此基础上,结合生态群落的演替进程,阐述了协同创新网络演替的5个阶段。最后,提出了加速协同创新网络演替的政策建议。